b
Discover
Models
Search
About
A Simple and Adaptive Learning Rate for FTRL in Online Learning with Minimax Regret of
Θ
(
T
2
/
3
)
\Theta(T^{2/3})
Θ
(
T
2/3
)
and its Application to Best-of-Both-Worlds
2 weeks ago
·
NeurIPS